Coumarin-Phosphine-Based Smart Probes for Tracking Biologically Relevant Metal Complexes: From Theoretical to Biological Investigations

Article abstract of DOI:10.1002/ejic.201501304

Ten metal-based complexes and associated ligands have been synthesized and characterized. One of the metal ligands is a coumarin-phosphine derivative, which displays tunable fluorescence properties. The fluorescence is quenched in the case of the free lig

Full text of DOI:10.1002/ejic.201501304

DOI: 10.1002/ejic.201501304
Full Paper
Trackable Therapeutic Agents
Coumarin-Phosphine-Based Smart Probes for Tracking
Biologically Relevant Metal Complexes: From Theoretical to
Biological Investigations
[a,b]
[c]
[d]
[d]
[d]
Lucile Dondaine,
Daniel Escudero, Moussa Ali, Philippe Richard, Franck Denat,
Ali Bettaieb,[ Pierre Le Gendre, Catherine Paul,
a,b]
[d]
[a,b]
Denis Jacquemin,*
[c,e]
[d]
[d]
Christine Goze,* and Ewen Bodio*
Abstract: Ten metal-based complexes and associated ligands than the emissive state and is responsible for the quenching of
II
II
have been synthesized and characterized. One of the metal li- fluorescence. For the Ru and Os complexes, other non-radia-
gands is a coumarin-phosphine derivative, which displays tun- tive channels involving the manifold of singlet and triplet ex-
able fluorescence properties. The fluorescence is quenched in cited states may play a role. The anti-proliferative properties of
the case of the free ligand and ruthenium and osmium com- all the compounds were evaluated in cancer cell lines (SW480,
plexes, whereas it is strong for the gold complexes and phos- HCT116, MDA-MB-231 and MCF-7); higher IC50 values were ob-
phonium derivatives. These trends were rationalized by theoret- tained for gold(I) complexes, with the free ligands being only
ical calculations, which revealed non-radiative channels involv- weakly cytotoxic.
ing a dark state for the free ligands that is lower in energy
Introduction
pass clinical trials due to their low efficiency in vivo and the
lack of understanding of their mechanism of action [target(s),
pharmacokinetics, stability]. This last criterion has become in-
creasingly crucial recently. One possible strategy for tackling
this issue is to accumulate data from a large number of biologi-
cal experiments. More recently, a growing community of chem-
ists have designed trackable therapeutic agents by tethering
an imaging probe to the therapeutic moiety. Such objects are
Metal-based complexes are one of the most studied and used
classes of anti-cancer drugs. Indeed, cisplatin and platinum de-
rivatives are present in more than 50 % of chemotherapeutic
anti-cancer cocktails.[ Owing to the heavy side-effects of plati-
num derivatives and to the acquired or intrinsic resistance of
some tumours to such treatments, numerous non-platinum
metal complexes have been investigated. Among them, ruthe-
nium and gold derivatives have provided the most promising
1]
[
3]
sometimes considered as theranostic agents, even though
their properties are quite far from Funkhouser's original defini-
[
2]
results. Nevertheless, most of these new complexes did not
[
4]
tion. This new strategy advantageously enables the real-time
tracking of drugs in vitro and/or in vivo without altering the
cells/animals treated. However, a recurring difficulty is ascertain-
ing whether the whole theranostic agent is actually tracked.
Indeed, proving the stability of compounds in complex biologi-
cal media is not straightforward, and if the integrity of the ther-
anostic is not conserved, it is only the probe moiety that is
imaged. One elegant strategy to circumvent this problem is to
[
[
[
a] Ecole Pratique des Hautes Etudes,
5014 Paris, France
www.ephe.fr
b] Univ. Bourgogne Franche-Comté, LIIC EA7269,
7
2
1000 Dijon, France
inserm-u866.u-bourgogne.fr
c] Chimie Et Interdisciplinarité, Synthèse, Analyse, Modélisation (CEISAM),
UMR CNRS 6320, BP 92208, Université de Nantes,
[
5]
2
, Rue de la Houssinière, 44322 Nantes Cedex 3, France
use smart probes.
E-mail: Denis.Jacquemin@univ-nantes.fr
www.univ-nantes.fr/jacquemin-d
Recently, we reported the synthesis, characterization and
biological evaluation of several gold(I) coumarin-phosphine
[
[
d] Institut de Chimie Moléculaire de l'Université de Bourgogne Franche-
Comté,
[
5]
complexes, CP-Au-1 and CP-Au-2 (Scheme 1). In this initial
study we showed that the coumarin-phosphine ligand presents
smart probe character: it displays strong fluorescence when
ICMUB UMR CNRS 6302,
9
avenue Alain Savary, BP 47870, 21078 Dijon Cedex, France
E-mail: christine.goze@u-bourgogne.fr
ewen.bodio@u-bourgogne.fr
www.icmub.fr
I
complexed to Au , but the fluorescence is dramatically
quenched in the case of decomplexation (Figure 1). This specific
photophysical behaviour enabled us to study its mechanism of
action.
e] Institut Universitaire de France (IUF),
1
rue Descartes, 75005 Paris Cedex 05, France
www.iufrance.fr
In particular, the chlorido-gold analogue CP-Au-1 (Scheme 1)
displayed promising results: a valuable cytotoxicity in human
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/ejic.201501304.
Eur. J. Inorg. Chem. 2016, 545–553
545
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 1. Synthesis of the phosphoniums CPMe+1 and CPMe+2 and of the complexes CP-Au-1, CP-Au-2, CP-AuS-1, CP-AuS-2, CP-Ru-1, CP-Ru-2, CP-Os-1, CP-
Os-2, CP-Pt-1 and CP-Pt-2.
vance. Our aim was not only to extend the scope of our previ-
ous smart theranostics, but to determine whether the couma-
rin-phosphine ligand could be used as a general tool for track-
ing transition-metal complexes in vitro.
First, the influence of the metal on the photophysical proper-
ties was investigated and particular attention was paid to their
ability to quench or otherwise the fluorescence. Theoretical cal-
culations were performed to rationalize these trends. Next, the
biological behaviour of the different complexes was evaluated.
I
Results and Discussion
Figure 1. Decrease in fluorescence upon the decomplexation of Au in CP-
Au-1 (reproduced with permission from our previous study ).
[
5]
Chemical Syntheses
cancer cells and a low toxicity in healthy zebrafish larvae. In the The coumarin-phosphine ligands CP-1 and CP-2 were synthe-
present study, we wished to generalize this initial investigation sized in two steps according to the procedure described by
[
6]
by varying the metal centre and, to this end, we selected ruthe- Pratt and co-workers (Scheme 1). The ligands were then
nium, osmium and platinum, which are all of biological rele- treated either with iodomethane or with metallic precursor
Eur. J. Inorg. Chem. 2016, 545–553
www.eurjic.org
546
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
(
[Au(tht)Cl],
[(p-cymene)RuCl ] ,
[(p-cymene)OsCl2]2
or inversion of one molecule). The Ru group adopts the classical
2
2
[
(cod)PtCl ]; tht = tetrahydrothiophene, cod = cycloocta-1,5- piano stool geometry.
2
[
5]
diene) to afford the corresponding phosphoniums CPMe+1
[
5]
and CPMe+2, and the metal complexes CP-Au-1, CP-Au-2,
CP-Ru-1, CP-Ru-2, CP-Os-1, CP-Os-2, CP-Pt-1 and CP-Pt-2 in ex-
cellent yields (Scheme 1).
Photophysical Studies
The complexation reactions were monitored by 31P NMR The photophysical characterization of the ligands CP-1 (tolyl)
3
1
1
spectroscopy. The P{ H} NMR spectra of the ligands display a and CP-2 (phenyl) and the different complexes was carried out
singlet at around –7 ppm, whereas the gold complexes display in dichloromethane and the data are summarized in Table 1.
a singlet at around 32 ppm, the ruthenium complexes a singlet
at around 22 ppm and the osmium complexes a singlet at
around –14 ppm (this negative chemical shift could be re-
garded as surprising but it is in fact expected for osmium–phos-
Table 1. Photophysical data[a] for the different compounds in CH
2
Cl
2
at 298 K.
b]
λ
abs
λ
em
ε
Φ [
f
Br
Br/Br(L)
–1
cm–1
–1
cm–1
]
[
nm] [nm]
[
M
]
[%]
[
M
[
7]
phine complexes). The platinum derivatives gave a slightly
more complex signal: the complexes display a single resonance
CP-1
348
430
441
432
432
433
432
432
430
440
430
433
428
430
433
25100
34300
25700
26200
26600
21800
59900
25000
23100
28600
24300
31000
28800
58300
3.1
91
83
83
5.8
9.5
4.8
5
98
92
95
7.5
1.3
4
778
1
40[
27
28
2
c]
CPMe+1
CP-Au-1
CP-AuS-1
CP-Ru-1
CP-Os-1
CP-Pt-1
CP-2
CPMe+2
CP-Au-2
CP-AuS-2
CP-Ru-2
CP-Os-2
CP-Pt-2
355
350
349
348
346
346
348
360
348
350
352
350
350
31213
21331
21746
1543
2071
2875
1250
22638
26312
23085
2325
374
3
1
1
in the P{ H} NMR spectra with platinum satellites separated
1
195
by a coupling constant J( Pt–P) of 3675 Hz, which indicates
the presence of the chlorine atoms trans to the phosphorus
[
8]
3
4
atoms and thus a cis geometry. Two additional gold(I) com-
plexes were synthesized by treating the chlorido–gold com-
plexes CP-Au-1 and CP-Au-2 with the in situ generated thiolate
1
18[c]
[
5]
of thioglucose tetraacetate. CP-AuS-1 and CP-AuS-2 were ob-
tained in high yields (Scheme 1).
21
18
2
0.3
2
All the compounds were fully characterized and the struc-
[
5]
[5]
tures of compounds CP-1, CP-2, CP-Au-1 and CP-Ru-2 were
confirmed by single-crystal X-ray diffraction analysis (Fig-
2332
[
a] λabs = wavelength of maximal absorption; λem = wavelength of maximal
[
9]
ure 2). CP-2 crystallizes in the P 1¯ space group and CP-Ru-2
crystallizes in the monoclinic P2 /n space group. The asymmet-
emission; ε = molar absorption coefficient; Φ = fluorescence quantum yield;
f
1
brightness, Br = εΦ ; Br(L) = brightness of ligand CP-1 or CP-2. [b] Reference:
f
= 0.97, λexc = 355 nm, in cyclohexane).[ [c] This
10]
diphenylanthracene (Φ
f
ric unit contains two independent molecules and one CH Cl
2
2
ratio, which illustrates the smart-probe character of the compounds, is not
relevant for the phosphonium derivatives because the “demethylation“ of
CPMe+1 and CPMe+2 is very unlikely.
solvate molecule. The two independent molecules of the CP-
Ru-2 cell share the same conformation (RMSD = 0.116 Å after
UV/Visible Studies
The ligands CP-1 and CP-2 each present an absorption maxi-
mum at 348 nm in CH Cl , which corresponds to the S band
2
2
1
(
see Theoretical Calculations section below), and have molar
–1
–1
absorption coefficients of 25100 and 25000
tively, which are characteristic of the coumarin signature.
M
cm , respec-
[
6]
A
small shoulder can be additionally observed for all the com-
pounds at the beginning of the S1 band (at around 300–
3
20 nm), which may correspond to the S transition or to a
2
vibronic coupling effect. These bands have readily been as-
signed to spin-allowed π–π* transitions. Alkylation of the phos-
phine induces a bathochromic shift of the absorption band of
I
II
II
about 10 nm. Complexation to Au , Ru or Os did not signifi-
cantly influence the molar absorption coefficients. A low-inten-
II
sity broad shoulder can be seen at 490 and 450 nm for the Ru
II
and Os complexes, respectively, which correspond to metal-
centred bands (see below). Finally, the presence of two
II
coumarin chromophores in the Pt complexes CP-Pt-1 and CP-
Pt-2 increases their absorption coefficients to 59900 and
–1
–1
5
8300
M
cm , respectively. These values are twice those ob-
tained for structures bearing a single coumarin fluorophore.
Exemplarily, the absorption and emission spectra of CP-2 and
CP-Ru-2 are represented in Figure 3 (see the Supporting Infor-
mation for the spectra of the other compounds).
Figure 2. ORTEP views of CP-2 (top) and CP-Ru-2 (bottom). The CH
2 2
Cl mol-
ecule is not shown for clarity and only one of the two independent molecules
[
5]
is represented (see our previous study for the ORTEP views of CP-1 and CP-
Au-1).
Eur. J. Inorg. Chem. 2016, 545–553
www.eurjic.org
547
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
(Φ = 91 % for CPMe+1 and 98 % for CPMe+2), because the
non-radiative channels are efficiently suppressed. Complexation
I
10
with Au , which is an inert d metal, induces the same phenom-
enon as alkylation of the phosphine, that is, blockage of the
lone pair on the phosphine. The resulting quantum yields of
the different Au complexes are therefore very high (between 83
and 95 %).
On the other hand, complexation with Ru , Os and PtII
strongly quenches the fluorescence of the coumarin. Even if the
quenching mechanism is also prevented in these compounds,
additional photophysical phenomena, which induce a decrease
in the fluorescence quantum yields, are involved. These mecha-
nisms could involve electron-transfer phenomena or triplet ex-
cited-state deactivation pathways that can be reached by inter-
system-crossing mechanisms. Indeed, the presence of metal
ions can induce the population of metal-to-ligand charge-trans-
II
II
3
fer ( MLCT) states and the eventual appearance of non-radiative
3
deactivation pathways involving metal-centred ( MC) states.
The brightness values are moderate to high. Therefore, all
the complexes, except CP-Os-2 and maybe CP-Ru-1, could most
probably be tracked in vitro in the micromolar range. The
“
smart-probe character” of the complexes can be evaluated by
the ratio of the brightness of the complexes to the brightness
of the corresponding coumarin-phosphine ligand CP-1 or CP-2
(
Table 1). As a result of the decrease in the quantum yields for
II
II
II
the Ru , Os and Pt complexes, this ratio is poor (0.3 to 4).
Thus, the decrease in fluorescence upon decomplexation would
not be sufficient to obtain an unambiguous in vitro interpreta-
tion. In contrast, all gold complexes display very good ratios,
Figure 3. Absorption (solid line) and emission (dashed line) spectra of CP-2
top) and CP-Ru-2 (bottom) in CH Cl at 298 K.
(
2
2
[
5]
from 18 to 28, thereby confirming previous observations.
Emission Properties
For all systems, a fluorescence band peaking at around 430 nm
can be observed, which corresponds to the emission of the
coumarin moiety. However, the rate of fluorescence significantly
Theoretical Calculations
depends upon the compound. First, the quantum yields of the To gain insights into the photophysical properties of the cou-
ligands CP-1 and CP-2 are relatively low (3.1 and 5 %, respec- marin-phosphine ligands and their complexes, we performed
tively). This quenching is primarily attributed in the photophys- DFT and time-dependent DFT (TD-DFT) calculations (see Com-
ics community to a photoinduced electron transfer (PET) from putational Details in the Exp. Sect.). Table 2 lists the TD-B3LYP
the phosphine to the fluorophore. As a result of our theoretical vertical excitation energies and oscillator strengths for the main
studies (see below), we herein propose that non-radiative chan- lowest singlet excited states of CP-1, CPMe+1, CP-Au-1 and CP-
nels involving a dark state are responsible for the observed Ru-1.
quenching of photoluminescence. The dark state is precisely
Two singlet excited states of mixed n /πcoum→π*coum charac-
P
the charge transfer that results from PET, and hence it further ter (see the specific excitations in Table 2 and the orbitals in-
validates our hypothesis. Alkylation of the phosphine results in volved in Figure 4, c) contribute to the main UV/Vis band of CP-
a strong enhancement of the fluorescence of the coumarin 1, which shows a peak experimentally at 350 nm. These excited
Table 2. TD-B3LYP/6-31G(d) vertical singlet electronic transition energies (in nm and eV) and oscillator strengths of CP-1, CPMe+1, CP-Au-1 and CP-Ru-1. The
main molecular orbital contributions are given in each case.
State
λ [nm/eV][a]
f [a.u.][a]
Character [Coeff.][a]
(H→L) n coum→π*coum [0.70]
(H→L) n coum→π*coum [0.70]
(H-1→L) πcoum →π*coum [0.69]
(H-1→L) πcoum →π*coum [0.69]
(H→L) πcoum→π*coum [0.70]
(H→L) πcoum→π*coum [0.70]
(H→L + 1) 4dt2g→4degπ*arene [0.63]
(H-1→L + 1) 4dt2g→4degπ*arene [0.42]
(H-3→L) πcoum→π*coum [0.70]
CP-1
S
1
375/3.31
0.447
0.705
0.442
0.311
0.827
0.941
0.007
0.015
0.843
P
π
P
π
3
70/3.35
326/3.80
27/3.79
S
2
n
n
P
P
3
CPMe+1
CP-Au-1
CP-Ru-1
S
1
S
1
S
1
S
4
385/3.21
348/3.57
550/2.25
455/2.73
350/3.55
S
10
[
a] PCM-TD-B3LYP/6-31G(d) values in CH
2 2
Cl are given in italics.
Eur. J. Inorg. Chem. 2016, 545–553 www.eurjic.org
548
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
states are theoretically located at 375 and 326 nm, and they predominantly lead to the population of the lowest-lying
gain intensity mainly from their πcoum→π*coum contribution. nPπ*coum state. A very small fluorescence rate (k ) is expected
fl
The inclusion of solvent effects leads to very small shifts of from this state (see its negligible oscillator strength in Table 3).
these excited states (ca. 0.01–0.04 eV, see Table 2). In the case Therefore, instead of radiative emission from the Sππ* state,
of the methylated CPMe+1 ligand and the CP-Au-1 complex, competing non-radiative decay channels leading to S will be
0
only the S state contributes to the experimental peaks ob- favoured for CP-1. These are presumably the ultimate reasons
1
served at 355 and 350 nm, respectively; good agreement with for the observed quenching of fluorescence in CP-1. We remark
the TD-B3LYP values (385 and 348 nm, respectively) is reached. that this theoretical modelling provides support for a non-radia-
For CP-Ru-1, the first nine singlets involve the RuCl –arene core, tive interpretation of the quenching of fluorescence rather than
2
see exemplarily the S and S states in Table 2. These states an intramolecular photoinduced electron transfer (PET) mecha-
1
4
possess very low oscillator strengths and they accordingly con- nism. To the best of our knowledge, only one previous theoreti-
tribute to the small broad tail extending up to 550 nm in the cal study proposed a similar non-radiative decay path for rhod-
recorded spectrum (see Figure 3). The πcoum→π*coum excitation amines.[11] In contrast, these non-radiative decay channels are
(
S ) is responsible for the main UV/Vis band of CP-Ru-1, and its not available for CPMe+1 or CP-Au-1 (their lowest excited state
10
position is again in very good agreement with the experimental is the bright Sππ* state, see Table 3 exemplarily for CPMe+1).
value.
Finally, other non-radiative mechanisms involving the manifold
of RuCl -based singlet and triplet states are opened up for CP-
2
Ru-1. These states are known to be involved in non-radiative
deactivation pathways involving crossings with the ground-
[
12]
state PES. These non-radiative channels might compete with
[
13]
PET.
Table 3. TD-B3LYP/6-31G(d) vertical singlet electronic emission energies (in
nm and eV) and oscillator strengths of CP-1 and CP-Au-1 for their Sππ* opti-
mized geometries. The main molecular orbital contributions are given in each
case.
State
λ [nm/eV]
f [a.u.]
Character [coeff.]
→π*coum [0.70]
coum→π*coum [0.70]
coum→π*coum [0.70]
CP-1
S
1
S
2
S
1
428/2.90
405/3.06
414/3.00
0.000
0.911
0.825
n
P
π
π
CPMe+1
Biological Studies
0
Figure 4. Main geometrical features of the S (a) and Sππ* (b) optimized ge-
ometries of CP-1. Kohn–Sham (KS) orbitals involved in the lowest excitations
of CP-1 in its S (c) and Sππ* (d) optimized geometries.
The anti-proliferative activity of the coumarin-derivatives was
tested on four human cancer cell lines of mammary (MDA-MB-
0
2
31 and MCF-7) or colon (SW480 and HCT116) origins (Table 4).
We next evaluated the emission properties and the quench-
ing of fluorescence in CP-1. Table 3 lists the TD-B3LYP emission
Table 4. IC50 values for the different coumarin derivatives against SW480,
HCT116, MDA-MB-231 and MCF-7 cells determined in MTS assays for 48 h.
energies for the optimized structures of the emissive state (Sππ*
)
of CP-1 and CPMe+1. The TD-B3LYP emission energies are in
reasonable agreement with experimental evidence. The main
IC50 [μM]
SW480
HCT116
MDA-MB-231 MCF-7
geometrical features of the S and S
optimized geometries
0
ππ*
CP-1
CPMe+1
CP-Au-1
160 ± 13
25.7 ± 0.3
42 ± 11
34 ± 14
75 ± 25
42 ± 14
insoluble
>200
75.6 ± 0.7
25.1 ± 0.1
25.2 ± 0.1
73 ± 7
122 ± 5
10.1 ± 0.2
38 ± 17
49.9 ± 0.3
74.9 ± 0.1
53 ± 14
>200
135 ± 14
28 ± 4
49 ± 1
50.1 ± 0.1
83 ± 15
50.0 ± 0.3
of CP-1 are highlighted in Figure 4, a,b. Relaxation of its Sππ*
potential energy surface (PES) leads to 1) a coplanarization of
coumarin and its adjacent tolyl ligand (see the φ1–2–3–4 dihedral
[a]
33 ± 14
33 ± 15
46 ± 5
74.9 ± 0.1
50.0 ± 0,1
[a]
CP-AuS-1[
CP-Ru-1
CP-Os-1
CP-Pt-1
CP-2
a]
angle in Figure 4, b) and 2) a twist of the Ptol moiety (compare
2
the φ5–6–7–8 and φ5–6–7–9 dihedral angles in Figure 4, a,b). In the
optimized S geometry, both the HOMO–1 and HOMO orbitals
0
104 ± 28
75.9 ± 0.8
26 ± 1
25.5 ± 0.1
49.4 ± 0.5
86 ± 13
74.8 ± 0.2
47 ± 5
25.3 ± 0.4
24.5 ± 0.3
25.3 ± 0.2
50 ± 1
>200
are delocalized over the phosphine and coumarin moieties (see
Figure 4, c). In contrast, in the optimized Sππ* geometry, these
orbitals evolve to well-separated nP and πcoum orbitals (see
HOMO–1 and HOMO in Figure 4, d). Consequently, pure
CPMe+2
CP-Au-2
CP-AuS-2
CP-Ru-2
CP-Os-2
CP-Pt-2
75.3 ± 0.5
25.3 ± 0.2
24.8 ± 0.1
49.8 ± 0.3
74.7 ± 0.1
>200
105 ± 27
49
nPπ*
coum
and πcoumπ*coum excited states are obtained for the
64 ± 26
58 ± 11
latter geometry (see Table 3). For this geometry, the dark
nPπ*coum state is located 0.16 eV below the emissive
πcoumπ*coum state. In this simplified photophysical picture for
[
a] Determined previously, see ref.[5]
An overview of the data indicated that the CP-1 and CP-2
CP-1, both the S and S states will be populated in the Franck– ligands are weakly cytotoxic and that the different metal-based
1
2
Condon region. From there, internal conversion processes will coumarin complexes are more potent towards the MDA-MB-231
Eur. J. Inorg. Chem. 2016, 545–553
www.eurjic.org
549 © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
cell line than towards the SW480, HCT116 and MCF-7 cell lines. and used as received. CP-1, CPMe+1, CP-Au-1, CP-AuS-1 and CP-
[
5,6]
2
were synthesized according to literature procedures.
All the
The complexes show substantial toxic effects with IC50 values
ranging from 24 to 106 μ . Among the metal complexes, the
gold(I) derivatives display the more promising properties (with
IC50 ranging from 24.5 to 50.1 μ ) and the general trend sug-
analyses were performed at the Plateforme d'Analyses Chimiques
et de Synthèse Moléculaire de l'Université de Bourgogne. The iden-
tities and purities (≥95 %) of the complexes were unambiguously
established by HRMS and NMR spectroscopy. The exact masses of
the synthesized complexes were determined with a Thermo LTQ
M
M
gests that the phenyl derivatives are a little more potent than
II
the tolyl derivatives. In contrast, the Os complexes present
1
13
31
Orbitrap XL spectrometer. H (300.13 MHz), C (75.47 MHz) and
(121.49 MHz) NMR spectra were recorded with a Bruker 300 Avance
P
higher IC50 values, which may be due to their lower solubility.
[
14]
Additionally, we note that the “X-type” ligand present in the III spectrometer. Chemical shifts (δ) are quoted in ppm relative to
gold complex (chlorido or thioglucose tetraacetate ligand) does TMS ( H and C) using the residual protonated solvent ( H) or the
1
13
1
13
31
not seem to influence the in vitro cytotoxic properties of the deuterated solvent ( C) as internal standards; 85 % H PO ( P) was
3 4
used as an external standard. The atom labelling used for the NMR
assignments is presented in Figure 5. IR spectra were recorded with
a Bruker Vector 22 FT-IR spectrophotometer (Golden Gate ATR).
gold complexes (CP-Au-1 and CP-AuS-1 vs. CP-Au-2 and CP-AuS-
). However, even if there is no notable difference in vitro, the
2
pharmacokinetic properties and the biodistribution of the com-
plexes will probably be different in vivo. Surprisingly, the phos-
phonium derivative CPMe+2 is significantly less cytotoxic than
its tolyl analogue CPMe+1 (up to 7.5 times in HCT116). Concern-
ing the Pt complexes CP-Pt-1 and CP-Pt-2, the IC50 values are
not really relevant, when determined, due to their poor water
solubility. CP-Pt-1 was even almost insoluble in pure DMSO.
Conclusions
We have designed, synthesized and characterized eight new
metal-based theranostic agents. The photophysical properties
were studied and compared. As highlighted in some of our pre-
vious work on BODIPY derivatives, gold derivatives gave the
most promising results and only the gold(I)–coumarin com-
plexes could be considered as smart theranostic com-
[
5,7,15]
pounds.
The quenching of the fluorescence of the ligands
II
and Ru complexes has been explained by theoretical calcula-
tions. Indeed, non-radiative channels involving a dark state,
which is lower in energy than the emissive state, are responsible
II
II
Figure 5. Arbitrary labelling of the ligand protons.
for the quenching of fluorescence. In the Ru and Os com-
plexes, other non-radiative channels involving the manifold of
singlet and triplet excited states might be operative and explain
the quenching. The anti-proliferative properties of all the com-
pounds were evaluated on SW480, HCT116, MDA-MB-231 and
6
[
(η -p-Cymene)(3-{4-(di-p-tolylphosphanyl)phenyl}-7-methoxy-
2
H-chromen-2-one)RuCl ] (CP-Ru-1): The reaction was carried out
2
under argon. 3-[4-(Di-p-tolylphosphino)phenyl]-7-methoxy-2H-
MCF-7 cell lines. The ligands appeared to be weakly cytotoxic, chromen-2-one (CP-1; 150 mg, 0.323 mmol) and [RuCl (p-cymene)]
2
2
whereas the complexes produced interesting IC50 values, espe- (99 mg, 0.162 mmol) were dissolved in degassed CH Cl (10 mL).
2
2
cially for the MDA cell line (with the exception of the poorly The resulting mixture was stirred at room temperature for 2 h. The
reaction was monitored by 31P NMR spectroscopy. Upon comple-
tion, the solvent was removed under reduced pressure. The ruthe-
nium complex CP-Ru-1 was isolated as a red powder (208 mg, 84 %
II
soluble Pt complex CP-Pt-1). The most interesting IC values
5
0
were obtained for the gold(I) complexes as well as the phos-
phonium derivatives. It is worth noting that even if the smart-
probe character is a bit less pronounced for the diphenylphos-
phine derivatives compared with the ditolylphosphines, their
anti-proliferative properties are improved.
As a consequence of the results described here, we are now
focusing our attention on gold(I) complexes by varying the sub- 8.2, J = 1.8 Hz, 4 H, H ), 7.43 (d, J = 8.4 Hz, 1 H, H ), 7.65 (dd,
stituent groups on the phosphorus atom combined with water-
solubilizing groups.
1
3
yield). H NMR (CDCl ): δ = 1.11 [d, J = 6.9 Hz, 6 H, CH(CH ) p-
3
H,H
3 2
cymene], 1.88 (s, 3 H, CH3 p-cymene), 2.36 (6 H, H ), 2.87 [hept,
j
3
JH,H = 6.9 Hz, 1 H, CH(CH ) p-cymene], 3.88 (s, 3 H, H ), 4.99 (dd,
3 2
a
3J
6
= 6.2, 4JH,P = 1.0 Hz, 2 H, CHAr p-cymene), 5.21 (d, JH,H
3
=
=
H,H
3
.2 Hz, 2 H, CH p-cymene), 6.83–6.89 (m, 2 H, H ), 7.18 (dd, J
Ar b,c H,H
4
3
H,P
i
H,H
d
3
4
J
H,H = 8.5, JH,P = 2.0 Hz, 2 H, H
), 7.69–7.78 (m, 5 H, He,h), 7.83–
): δ = 18.0, 21.5, 22.0,
f
1
7.90 (m, 2 H, H
) ppm. 13C{ H} NMR (CDCl
g
3
3
1
0.4, 56.0, 87.4 (d, JC,P = 5.5 Hz), 89.0 (d, JC,P = 3.1 Hz), 96.0, 100.6,
11.2 (d, JC,P = 3.1 Hz), 113.0, 113.4, 123.9 (d, JC,P = 1.1 Hz), 127.7
(
4
d, JC,P = 9.9 Hz), 128.9 (d, JC,P = 10.3 Hz), 129.3, 130.6 (d, JC,P
7.2 Hz), 134.4 (d, JC,P = 46.2 Hz), 134.5 (d, JC,P = 9.9 Hz), 134.5 (d,
=
Experimental Section
General: All the reactions were carried out under purified argon
using Schlenk techniques. Solvents were dried and distilled under
argon before use. All other reagents were commercially available
JC,P = 9.4 Hz), 136.6 (d, JC,P = 2.8 Hz), 140.6 (d, J
= 2.8 Hz),
C,P
31
1
140.9, 155.6, 160.8, 163.1 ppm. P{ H} NMR (CDCl ): δ = 23.2 ppm.
3
–
1
–1
UV/Vis (CH Cl ): λ
(ε) = 348 nm (26600
M
cm ). IR: ν = 1653
2
2
max
˜
Eur. J. Inorg. Chem. 2016, 545–553
www.eurjic.org
550
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
–
1
143.1, 155.9, 160.5, 163.8 ppm. 31P{ H} NMR (CDCl ): δ = 21.2 ppm.
1
(
ν
) cm .HRMS (ESI): calcd. for C H ClO PRu 735.13712 [M –
C=O
+
40 39
3
3
–1
–1
Cl] ; found 735.13984.
UV/Vis (CH Cl ): λ (ε) = 360 (23100
max
νC=O) cm . HRMS (ESI): calcd. for C H O P 451.14576 [M – I] ;
M
cm ). IR: ν˜ = 1653
2
2
–
1
+
+
(
29
24 3
6
[
2
(η -p-Cymene)(3-{4-(di-p-tolylphosphanyl)phenyl}-7-methoxy-
H-chromen-2-one)OsCl ] (CP-Os-1): The reaction was carried out
found 451.14421.
2
under argon. CP-1 (132 mg, 0.285 mmol) and [OsCl (p-cymene)]2 [(3-{4-(Diphenylphosphanyl)phenyl}-7-methoxy-2H-chromen-2-
2
(
113 mg, 0.142 mmol) were dissolved in degassed CH Cl (10 mL).
one)AuCl] (CP-Au-2): The reaction was carried out under argon. 3-
[4-(Diphenylphosphino)phenyl]-7-methoxy-2H-chromen-2-one (CP-
2; 200 mg, 0.46 mmol) and [Au(tht)Cl] (147 mg, 0.46 mmol) were
dissolved in degassed CH Cl (12 mL). The resulting mixture was
2
2
The resulting mixture was stirred at room temperature for 2 h. The
3
1
reaction was monitored by P NMR spectroscopy (242.9 MHz,
00 K). Upon completion, the solvent was removed under reduced
pressure. The osmium complex CP-Os-1 was isolated as pale-brown
3
2
2
stirred at room temperature for 2 h. The reaction was monitored by
1
3
31
powder (193 mg, 79 % yield). H NMR (CDCl ): δ = 1.16 [d, J
=
P NMR spectroscopy. Upon completion, the solvent was removed
3
H,H
6
.9 Hz, 6 H, CH(CH ) p-cymene], 1.98 (s, 3 H, CH p-cymene), 2.36
under reduced pressure. The gold complex CP-Au-1 was isolated as
3
2
3
3
1
(6 H, H ), 2.76 [hept, J = 6.9 Hz, 1 H, CH(CH ) p-cymene], 3.88
a yellow powder (285 mg, 93 % yield). H NMR (CDCl ): δ = 3.90 (s,
j
H,H
3 2
3
3
4
3
4
(
s, 3 H, H ), 5.18 (d, JH,H = 5.7 Hz, 2 H, CHAr p-cymene), 5.42 (d,
JH,H = 5.7 Hz, 2 H, CH p-cymene), 6.83–6.88 (m, 2 H, Hb,c), 7.18
3 H, H ), 6.87 (d, J = 2.4 Hz, 1 H, H ), 6.90 (dd, J = 8.4, J
=
a
a
H,H
b
H,H
H,H
3
3
2.4 Hz, 1 H, H ), 7.43–7.63 (m, 13 H, H
), 7.80 (dd, J
= 8.4,
Ar
c
d,g,h,i,j
H,H
3
4
3
4
13
1
(
dd, J = 8.2, J = 1.8 Hz, 4 H, H ), 7.43 (d, J = 8.4 Hz, 1 H,
JH,P = 2.1 Hz, 2 H, H ), 7.84 (s, 1 H, H ) ppm. C{ H} NMR (CDCl ):
H,H
H,P
i
H,H
f e 3
H ), 7.62–7.70 (m, 6 H, H ), 7.77–7.85 (m, 3 H, H ) ppm. 13C{ H} δ = 56.0, 100.6, 113.3 (d, JC,P = 11.9 Hz), 123.1 (d, JC,P = 1.2 Hz),
1
d
f,h
e,g
NMR (CDCl ): δ = 18.0, 21.5, 22.4, 30.2, 56.0, 80.1 (d, J = 5.1 Hz),
124.8, 128.2, 128.7 (d, JC,P = 62.6 Hz), 129.0, 129.3 (d, JC,P = 15.5 Hz),
129.4 (d, JC,P = 12.1 Hz), 132.2 (d, JC,P = 2.6 Hz), 134.3 (d, JC,P
14.0 Hz), 138.9 (d, J = 2.6 Hz), 141.4, 155.8, 160.6, 163.4 ppm.
3
C,P
8
1
1
9
1
0.3 (d, JC,P = 2.5 Hz), 88.6, 100.6, 103.5 (d, JC,P = 4.0 Hz), 113.0,
13.4, 123.9, 127.6 (d, JC,P = 10.4 Hz), 128.8 (d, JC,P = 10.4 Hz), 129.3,
=
C,P
31
1
30.1 (d, JC,P = 54.4 Hz), 134.0 (d, JC,P = 52.6 Hz), 134.6 (d, JC,P
=
P{ H} NMR (CDCl ): δ = 32.7 ppm. UV/Vis (CH Cl ): λ
(ε) = 350
3
2
2
max
–
1
–1
–1
.8 Hz), 134.7 (d, JC , P = 9.8 Hz), 136.6 (d, JC , P = 2.7 Hz),
(25700
M
cm ). IR: ν˜ = 326.83 (νAu–Cl), 1653 (νC=O) cm . HRMS
40.7 (d, JC,P = 2.1 Hz), 140.9, 155.6, 160.8, 163.0 ppm. 31P{ H} (ESI): calcd. for C
1
H
AuClO PNa 691.04746 [M + Na] ; found
3
+
28
21
NMR (CDCl ): δ = –14.1 ppm. UV/Vis (CH Cl ): λ
(ε) = 346 nm
691.04815.
3
2
–1
2
max
–1
–1
(26600
M
cm ). IR: ν˜ = 1653 (ν ) cm . HRMS (ESI): calcd. for
C=O
+
C H ClO POs 825.194 [M – Cl] ; found 825.19609.
[(3-{4-(Diphenylphosphanyl)phenyl}-7-methoxy-2H-chromen-2-
one)gold(I)(thio-ꢀ- -glucose tetraacetate)] (CP-AuS-2): The reac-
40
39
3
D
[
Bis(3-{4-(di-p-tolylmethylphosphanyl)phenyl}-7-methoxy-2H-
tion was carried out under argon. [(3-{4-(Diphenylphosphan-
chromen-2-one)PtCl ] (CP-Pt-1): The reaction was carried out un-
yl)phenyl}-7-methoxy-2H-chromen-2-one)AuCl] (CP-Au-2; 128mg,
2
der argon. 3-[4-(Di-p-tolylphosphino)phenyl]-7-methoxy-2H-
0.193 mmol) was dissolved in degassed CH Cl (10 mL). 1-Thio-ꢀ-D-
2 2
chromen-2-one (CP-1; 150 mg, 0.323 mmol) and cyclooctadiene-
glucose tetraacetate (70.3 mg, 0.193 mmol) dissolved in distilled
platinum(II) dichloride (60.4 mg, 0.161 mmol) were dissolved in de- acetone (5 mL), NaOH (7.7 mg, 0.193 mmol) and four drops of water
gassed CH Cl (10 mL). The resulting mixture was stirred at room were introduced into a Schlenk tube. The reaction mixture was
temperature for 2 h. The reaction was monitored by P NMR spec- stirred for 10 min at room temperature in the dark. This mixture
2
2
3
1
troscopy (242.9 MHz, 300 K). Upon completion, the solvent was
removed under reduced pressure. The platinum complex CP-Pt-1
was isolated as a yellow powder (360 mg, 92 % yield). H NMR
was slowly added to the first one at 0 °C and then stirred for 3 h
at room temperature in the dark. The reaction mixture was filtered
to remove salts and the solvent was removed under reduced pres-
1
(
6
(
CDCl ): δ = 2.34 (12 H, H ), 3.89 (s, 6 H, H ), 6.82–6.90 (m, 4 H, Hb,c), sure. The complex CP-AuS-2 was isolated as a yellow powder
3
j
a
3
1
.99 (br. d, J = 7.1 Hz, 8 H, H ), 7.34–7.58 (m, 18 H, Hd,f,g,h), 7.81
(177 mg, 92 % yield). H NMR (CDCl ): δ = 1.90 [s, 3 H, CH C(O)],
H,H
i
3
3
13
1
br. s, 2 H, H ) ppm. C{ H} NMR (CDCl ): δ = 21.4, 55.8, 100.4, 113.1
1.97 [s, 3 H, CH C(O)], 2.02 [s, 3 H, CH C(O)], 2.05 [s, 3 H, CH C(O)],
e
3
3 3 3
3
(d, JC,P = 20.5 Hz), 123.1, 126.4, 127.4, 128.6, 128.7, 128.8, 129.4,
3.74–3.80 (m, s, CH sugar), 3.90 (s, 3 H, H ), 4.13 (dd, J = 12.3,
2 a H,H
31
1
2
3
3
1
34.2, 135.0, 135.1, 141.2, 141.3, 155.5, 160.6, 163.0 ppm. P{ H}
JH,H = 2.6 Hz, 1 H, CH sugar), 4.22 (dd, J = 12.3, J = 4.8 Hz,
2 H,H H,H
1
4
NMR (CDCl ): δ= 12.6 (s+d, J = 3675 Hz) ppm. UV/Vis (CH Cl ):
λmax (ε) = 346 nm (59900
(
1
1 H, CHCH sugar), 5.0–5.2 (m, 4 H, CH sugar), 6.86 (d, J = 2.4 Hz,
2 H,H
3
P, P t
2
2
–
1
–1
–1
3
4
M
cm ). IR: ν˜ = 1653 (ν ) cm . HRMS 1 H, H ), 6.89 (dd, J = 8.4, J = 2.4 Hz, 1 H, H ), 7.45–7.65 (m,
C=O
b
H,H
H,H
c
+
3
4
ESI): calcd. for C60H50ClO P Pt 1158.24183 [M – Cl] ; found
13 H, Hd,g,h,i,j), 7.83 (dd, J = 8.4, J = 2.0 Hz, 2 H, H ), 7.88 (s,
H,H H,P f
3
2
13
1
158.24245.
1 H, H ) ppm. C{ H} NMR (CDCl ): δ = 20.7, 20.7, 20.7, 21.1, 55.9,
e
3
7
4.2, 74.8, 100.5, 113.1 (d, JC,P = 1.3 Hz), 123.1 (d, JC,P = 1.2 Hz),
3
-[4-(Methyldiphenylphosphonium)phenyl]-7-methoxy-2H-
128.9, 129.1, 129.3, 129.3 (d, JC,P = 11.4 Hz), 131.9 (d, JC,P = 2.8 Hz),
134.3 (d, JC,P = 13.6 Hz), 138.6, 141.3, 155.6, 160.5, 163.3, 169.6,
chromen-2-one Iodide (CPMe+2) : The reaction was carried out
under argon. CP-1 (100 mg, 0.229 mmol) and iodomethane (49 mg,
0
170.3, 170.8 ppm. 31P{ H} NMR (CDCl ): δ = 36.2 ppm. UV/Vis
1
3
–
1
–1
.343 mmol) were dissolved in degassed CH Cl (7 mL). The result- (CH Cl ): λ
(ε) = 350 (24300
M
cm ). IR: ν = 370.55 (νAu–Sucre),
2
2
2
2
max
˜
ing mixture was stirred at room temperature for 2 h. The reaction 1653 (νC=O) cm . HRMS (ESI): calcd. for C42H40O12SPAuNa+
–1
was monitored by 31P NMR spectroscopy. Upon completion, the
solvent was removed under reduced pressure. The phosphonium
CPMe+2 was isolated as a yellow powder (127 mg, 96 % yield). H
NMR (CDCl ): δ = 3.16 (d, J = 13.4 Hz, 3 H, P-CH ), 3.89 (s, 3 H,
H ), 6.82 (d, J
+
1019.15413 [M + Na] ; found 1019.15672.
1
6
[(η -p-Cymene)(3-{4-(diphenylphosphanyl)phenyl}-7-methoxy-
2H-chromen-2-one)RuCl ] (CP-Ru-2): The reaction was carried out
under argon. 3-[4-(Diphenylphosphino)phenyl]-7-methoxy-2H-
chromen-2-one (CP-2; 150 mg, 0.344 mmol) and [RuCl (p-cymene)]
2
(105 mg, 0.172 mmol) were dissolved in degassed CH Cl (10 mL).
2
3
H,P
3
2
4
3
4
= 2.4 Hz, 1 H, H ), 6.88 (dd, J
= 8.6, J
=
a
H,H
b
H,H
H,H
3
2
.4 Hz, 1 H, H ), 7.61–7.86 (m, 13 H, Hd,g,h,i,j), 8.11 (dd, J
= 8.7,
c
H,H
2
4
13
1
JH,P = 3.1 Hz, 2 H, H ), 8.20 (s, 1 H, H ) ppm. C{ H} NMR (CDCl ):
f
e
3
2 2
δ = 11.6 (d, JC,P = 57.3 Hz), 56.1, 100.6, 113.2 (d, JC,P = 14.0 Hz),
The resulting mixture was stirred at room temperature for 2 h. The
reaction was monitored by 31P NMR spectroscopy. Upon comple-
tion, the solvent was removed under reduced pressure. The ruthe-
nium complex CP-Ru-2 was isolated as a red powder (222 mg, 87 %
1
1
17.8 (d, JC,P = 89.9 Hz), 118.5, 119.7, 121.5 (d, JC,P = 1.2 Hz), 130.3,
30.3 (d, JC,P = 13.2 Hz),130.6, 130.8, 133.4 (d, JC,P = 10.9 Hz), 133.6
(d, JC,P = 11.4 Hz), 135.4 (d, JC,P = 2.8 Hz), 142.2 (d, JC,P = 3.0 Hz),
Eur. J. Inorg. Chem. 2016, 545–553
www.eurjic.org
551
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1
3
yield). H NMR (CDCl ): δ = 1.10 [d, J = 7.0 Hz, 6 H, CH(CH ) p-
strument response. The fluorescence quantum yields (Φ ) were cal-
culated from Equation (1)
3
H,H
3 2
F
3
cymene], 1.88 (s, 3 H, CH p-cymene), 2.84 [hept, J = 7.0 Hz, 1
H, CH(CH ) p-cymene], 3.87 (s, 3 H, H ), 5.02 (d, J = 6.2 Hz, 2 H,
CHAr p-cymene), 5.21 (d, J = 6.2 Hz, 2 H, CHAr p-cymene), 6.80–
.89 (m, 2 H, Hb,c), 7.33–7.45 (m, 7 H, Hd,g, H ), 7.67 (dd, J = 8.4,
= 2.1 Hz, 2 H, H ), 7.75 (s, 1 H, H ), 7.81–7.94 (m, 6 H, H ) ppm.
C{ H} NMR (CDCl ): δ = 17.9, 22.0, 30.4, 55.8, 87.3 (d, J = 5.4 Hz),
3
H,H
3
3
2
a
H,H
3
H,H
3
6
Ph
H,H
4
J
H,P
1
f e Ph
1
3
3
C,P
8
1
9.1 (d, JC,P = 3.3 Hz), 96.2, 100.6, 111.2 (d, JC,P = 3.3 Hz), 113.0,
13.3, 123.8 (d, JC,P = 1.0 Hz), 127.8 (d, JC,P = 10.4 Hz), 128.2 (d,
(
1)
JC,P = 9.9 Hz), 129.3, 130.4 (d, JC,P = 2.1 Hz), 133.7 (d, JC,P = 45.7 Hz),
1
9
34.0 (d, JC,P = 45.7 Hz), 134.5 (d, JC,P = 9.4 Hz), 134.7 (d, JC,P
=
in which Φ and Φ are the fluorescence quantum yields of the
F
FR
.4 Hz), 136.9 (d, JC,P = 2.8 Hz), 141.0, 155.6, 160.8, 163.1 ppm.
compound and the reference, respectively, A(λ ) and A (λ ) are the
E
R
E
3
1
1
P{ H} NMR (CDCl ): δ = 23.9 ppm. UV/Vis (CH Cl ): λ
(ε) = 348
3
2
2
max
absorbances at the excitation wavelength of the compound and
the reference, respectively, and n is the refractive index of the me-
dium, and I and I are the fluorescent intensities of the compound
–1
–1
–1
(31000
M
cm ). IR: ν˜ = 1653 (ν ) cm . HRMS (ESI): calcd. for
C=O
+
C H Cl O PRu 707.10577 [M – Cl] ; found 707.10608; calcd. for
C H Cl O PRuNa 765.06412 [M + Na] ; found 765.06388.
3
8
35
2 3
F
FR
+
3
8
35
2
3
and the reference, respectively. 9,10-Diphenylanthracene (Φ = 0.97
F
[
10]
6
in cyclohexane) was used as standard.
In all the Φ determina-
F
[
2
(η -p-Cymene)(3-{4-(diphenylphosphanyl)phenyl}-7-methoxy-
H-chromen-2-one)OsCl ] (CP-Os-2): The reaction was carried out
tions, correction for the solvent refractive index (η) was applied.
2
under argon. CP-2 (150 mg, 0.344 mmol) and [OsCl (p-cymene)]2 X-ray Structures: Suitable crystals of CP-2 and CP-Ru-2 were se-
2
(
136 mg, 0.172 mmol) were dissolved in degassed CH Cl (12 mL).
lected and mounted on a Bruker APEX-II CCD diffractometer. The
2
2
crystals were kept at 115 K during data collection. Using Olex2,[
16]
The resulting mixture was stirred at room temperature for 2 h. The
reaction was monitored by P NMR spectroscopy. Upon comple-
3
1
[17]
the structures were solved by direct methods using the SHELXT
tion, the solvent was removed under reduced pressure. The osmium structure solution program and refined by least squares minimiza-
[
18]
complex CP-Os-2 was isolated as a pale-brown powder (215 mg,
tion using the SHELXL refinement package. Except for the minor
component of a disordered isopropyl group in CP-Ru-2 (C47a and
C48a), all non-hydrogen atoms were refined by using anisotropic
thermal parameters. Hydrogen atoms were refined by using a riding
model.
1
3
7
5 % yield). H NMR (CDCl ): δ = 1.16 [d, J = 6.9 Hz, 6 H, CH(CH )
3 H,H 3 2
3
p-cymene], 1.99 (s, 3 H, CH p-cymene), 2.74 [hept, J = 6.9 Hz,
1
H, CHAr p-cymene), 5.41 (d, JH,H = 5.9 Hz, 2 H, CHAr p-cymene),
6
8
7
5
3
H,H
3
H, CH(CH ) p-cymene], 3.88 (s, 3 H, H ), 5.20 (d, J = 5.9 Hz, 2
3 2 a H,H
3
3
.83–6.89 (m, 2 H, Hb,c), 7.33–7.41 (m, 6 H, H , H ), 7.43 (d, J
=
g
Ph
H,H
CCDC 1420552 (for CP-2) and 1420553 (for CP-Ru-2) contain the
supplementary crystallographic data for this paper. These data can
be obtained free of charge from The Cambridge Crystallographic
Data Centre.
3
4
.4 Hz, 1 H, H ), 7.68 (dd, J = 8.4, J = 2.1 Hz, 2 H, H ), 7.75–
.88 (m, 7 H, H , H ) ppm. C{ H} NMR (CDCl ): δ = 18.0, 22.3, 30.2,
d
H,H
3
H,P
f
1
1
e
Ph
3
6.0, 80.1 (d, JC,P = 5.0 Hz), 80.6 (d, JC,P = 2.8 Hz), 88.8, 100.6, 103.5
(d, JC,P = 3.9 Hz), 113.1, 113.4, 123.8 (d, JC,P = 1.1 Hz), 127.8 (d, JC,P
=
1
1
9
1
0.4 Hz), 128.1 (d, JC,P = 10.2 Hz), 129.3, 130.5 (d, JC,P = 2.2 Hz),
Computational Details: The geometries of the singlet ground state
(S
0
) and the emissive state (Sππ*) were optimized at the B3LYP[
19]
33.2 (d, JC,P = 52.3 Hz), 133.6 (d, JC,P = 52.3 Hz), 134.6 (d, JC,P
=
.4 Hz), 134.9 (d, JC,P = 9.4 Hz), 136.8 (d, J = 2.4 Hz), 141.0, 155.6,
and TD-B3LYP levels of theory, respectively, for CP-1, CPMe+1, CP-
Au-1 and CP-Ru-1. For these optimizations the 6-31G(d) atomic ba-
C,P
60.8, 163.1 ppm. 31P{ H} NMR (CDCl ): δ = –13.3 ppm. UV/Vis
1
3
–1
–
1
–1
(CH Cl ): λ
(ε) = 350 (28800
M
cm ). IR: ν˜ = 1653 (ν ) cm . sis set was selected. Relativistic effects were included for the Ru
2
2
max
C=O
+
HRMS (ESI): calcd. for C H Cl O POs – Cl 797.1628 [M – Cl] ; found
and Pt atoms by using the ECP-28-mwb and ECP-60-mwb Stuttgart/
3
8
35
2 3
[
20]
7
97.16421.
Dresden pseudopotentials.
The nature of the stationary points
was confirmed by computing the Hessian at the same level of the-
ory. The UV/Vis absorption spectrum was obtained by calculating
[
Bis(3-{4-(diphenylphosphanyl)phenyl}-7-methoxy-2H-chrom-
en-2-one)PtCl ] (CP-Pt-2): The reaction was carried out under ar-
gon. CP-2 (150 mg, 0.344 mmol) and cyclooctadieneplatinum(II) di-
2
vertical singlet electronic excitations using TD-B3LYP at the S ge-
0
ometries. Some of the TD-B3LYP calculations were performed in
2
chloride (64.3 mg, 0.172 mmol) were dissolved in degassed CH Cl2
[21]
CH Cl solution using the polarization continuum model (PCM).
2
2
(10 mL). The resulting mixture was stirred at room temperature for
The emission excitation energies were obtained by computing the
TD-B3LYP excitations for the optimized Sππ* geometries. All the TD-
B3LYP calculations were performed with the same basis set as in
the optimizations. All calculations were performed by using the
3
1
2
h. The reaction was monitored by P NMR spectroscopy. Upon
completion, the solvent was removed under reduced pressure. The
platinum complex CP-Pt-2 was isolated as a yellow powder (360 mg,
9
2
7
H ) ppm. C{ H} NMR (CDCl ): δ = 56.0, 100.5, 113.2 (d, J =
1
5
1
4
2 % yield). H NMR (CDCl ): δ = 3.89 (s, 6 H, H ), 6.83 (d, J
=
3
3
a
H,H
[22]
Gaussian09 program package.
4
.4 Hz, 1 H, H ), 6.87 (dd, J = 8.6, J = 2.4 Hz, 1 H, H ), 7.18–
b
H,H
H,H
c
.26 (m, 8 H, H ), 7.32–7.80 (m, 22 H, Hd,f,g,h,j), 7.83 (br. s, 2 H,
i
1
3
1
e
3
C,P
Acknowledgments
2.4 Hz), 123.2, 127.8 (pseudo t, J = 5.6 Hz), 128.2 (pseudo t, J =
.6 Hz), 129.6, 131.2, 134.5 (pseudo t, J = 5.4 Hz), 135.2 (pseudo t,
The authors are grateful to the Centre National de la Recherche
Scientifique (CNRS) (ICMUB, UMR CNRS 6302) for financial sup-
port. Support was also provided by the Université de Bourgo-
3
1
1
J = 5.2 Hz), 137.4, 141.4, 155.6, 160.7, 163.2 ppm. P{ H} NMR
1
(CDCl ): δ = 13.5 (s+d, J = 3675 Hz) ppm. UV/Vis (CH Cl ): λ
3
P, P t
–1
2
2
max
–
1
–1
(ε) = 350 (58300
M
cm ). IR: ν˜ = 1653 (ν ) cm . HRMS (ESI): gne and the Conseil Régional de Bourgogne through the 3MIM
C=O
+
calcd. for C H ClO P Pt 1102.17921 [M – Cl] ; found 1102.17991.
integrated project (Marquage de Molécules par les Métaux pour
l'Imagerie Médicale) and PARI SSTIC n°1 and n°6. Dr. Fanny Pic-
quet, Marie José Penouilh and Dr. Michel Picquet are acknowl-
56
42
6 2
Photophysical Characterization: UV/Vis absorption spectra were
recorded with a JASCO V630BIO spectrometer. The steady-state
fluorescence emission spectra were recorded with a JASCO FP8560 edged for technical support. E. B. thanks COST Action TD1004
spectrofluorimeter. All fluorescence spectra were corrected for in- for financial support. D. E. acknowledges the European Research
Eur. J. Inorg. Chem. 2016, 545–553
www.eurjic.org
552
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[
9] Crystal data for CP-2: C28
H
21
O
3
P (M = 436.42 g/mol), triclinic, space
Council (ERC) (Marches-278845) and the Région des Pays de la
Loire for his postdoctoral grant. D. J. acknowledges the Euro-
pean Research Council (ERC) and the Région des Pays de la
Loire for financial support within the framework of a starting
grant (Marches-278845) and the LUMOMAT project, respec-
tively. This research used the resources of the GENCI-CINES/ID-
RIS, the Centre de Calcul Intensif des Pays de Loire (CCIPL) and
a local Troy cluster.
group P1¯ (no. 2), a = 6.5730(4), b = 10.3480(6), c = 16.5877(10) Å, α =
3
8
1
4.354(2), ꢀ = 81.001(2), γ = 87.0180(10)°, V = 1108.19(11) Å , Z = 2, T =
15 K, μ(Mo-K ) = 0.152 mm , Dcalcd. = 1.308 g/cm , 34563 reflections
α
–1
3
measured (4.994° ≤ 2θ ≤ 55.23°), 5143 unique (Rint = 0.0223, Rsigma
.0126), which were used in all calculations. The final R value was 0.0322
I > 2σ(I)] and wR was 0.0879 (all data). Crystal data for CP-Ru-2:
(M = 1569.11 g/mol), monoclinic, space group P2 /n
=
0
1
[
2
C
77
H71Cl
6
O
6
P
2
Ru
2
1
(
no. 14), a = 12.3001(6), b = 25.1497(12), c = 22.1034(12) Å,
3
ꢀ = 99.4300(10)°, V = 6745.2(6) Å , Z = 4, T = 115 K, μ(Mo-K ) =
α
–
1
3
0
2
.788 mm , Dcalcd. = 1.545 g/cm , 88328 reflections measured (3.726° ≤
θ ≤ 55.168°), 15493 unique (Rint = 0.0527, Rsigma = 0.0415), which were
used in all calculations. The final R
1
value was 0.0377 [I > 2σ(I)] and
Keywords: Bioinorganic chemistry · Theranostic agents ·
Antitumor agents · P ligands · Fluorescent probes
wR was 0.0889 (all data). CCDC 1420552 (for CP-2) and 1420553 (for CP-
2
Ru-2) contain the supplementary crystallographic data for this paper.
These data are provided free of charge by The Cambridge Crystallo-
graphic Data Centre.
[
[
10] A. M. Brouwer, Pure Appl. Chem. 2011, 83, 2213–2228.
11] M. Savarese, U. Raucci, C. Adamo, P. A. Netti, I. Ciofinic, N. Rega, Phys.
Chem. Chem. Phys. 2014, 16, 20681–20688.
12] a) D. Escudero, E. Heuser, R. J. Meier, M. Schäferling, W. Thiel, E. Holder,
Chem. Eur. J. 2013, 19, 15639–15644; b) J.-L. Heully, F. Alary, M. Boggio-
Pasqua, J. Chem. Phys. 2009, 131, 184308.
[
1] a) A. V. Klein, T. W. Hambley, Platinum-Based Anticancer Agents, in: Ligand
Design in Medicinal Inorganic Chemistry (Ed.: T. Storr), Wiley, Chichester,
UK, 2014, p. 9–10; b) T. C. Johnstone, K. Suntharalingam, S. J. Lippard,
Phil. Trans. R. Soc. A 2015, 373:20140185.
[
[
2] a) N. Muhammad, Z. Guo, Curr. Opin. Chem. Biol. 2014, 19, 144–153; b)
C. G. Hartinger, P. J. Dyson, Chem. Soc. Rev. 2009, 38, 391–401; c) S.
Komeda, A. Casini, Curr. Top. Med. Chem. 2012, 12, 219–235; d) A. F. A.
Peacock, P. J. Sadler, Chem. Asian J. 2008, 3, 1890–1899; e) G. Gasser,
I. Ott, N. Metzler-Nolte, J. Med. Chem. 2011, 54, 3–25; f) E. R. Tiekink,
Inflammopharmacology 2008, 16, 138–142; g) S. Nobili, E. Mini, I. Landini,
C. Gabbiani, A. Casini, L. Messori, Med. Res. Rev. 2010, 30, 550–580; h) I.
Ott, Coord. Chem. Rev. 2009, 253, 1670–1681; i) B. Bertrand, A. Casini,
Dalton Trans. 2014, 43, 4209–4219; j) W. Fiskus, N. Saba, M. Shen, M.
Ghias, J. Liu, S. Das Gupta, L. Chauhan, R. Rao, S. Gunewardena, K.
Schorno, C. P. Austin, K. Maddocks, J. Byrd, A. Melnick, P. Huang, A. Wiest-
ner, K. N. Bhalla, Cancer Res. 2014, 74, 2520–2532; k) for examples of
clinical trials in progress in 2014–2015, see: https://clinicaltrials.gov/ct2/
show/NCT01747798, https://clinicaltrials.gov/ct2/show/NCT01419691
and http://www. cancer.gov/clinicaltrials/search/view?cdrid=743608&
version=HealthProfessional#StudyIdInfo_CD0000743608.
[
13] A. R. Santos, D. Escudero, L. González, G. Orellana, Chem. Asian J. 2015,
1
0, 622–629.
[
14] “X-type” ligands are ligands that donate one electron to the metal and
accept one from it when using the neutral ligand method of electron
counting.
15] a) S. Tasan, C. Licona, P.-E. Doulain, C. Michelin, C. P. Gros, P. Le Gendre,
P. D. Harvey, C. Paul, C. Gaiddon, E. Bodio, J. Biol. Inorg. Chem. 2015, 20,
[
1
43–154; b) P. D. Harvey, S. Tasan, C. P. Gros, C. H. Devillers, P. Richard, P.
Le Gendre, E. Bodio, Organometallics 2015, 34, 1218–1227.
16] O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard, H. Puschmann,
J. Appl. Crystallogr. 2009, 42, 339–341.
[
[17] G. M. Sheldrick, Acta Crystallogr., Sect. A 2015, 71, 3–8.
[18] G. M. Sheldrick, Acta Crystallogr., Sect. A 2008, 64, 112–122.
[19] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652; b) C. T. Lee, W. T.
Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785–789.
[
3] R. Kumar, W. S. Shin, K. Sunwoo, W. Y. Kim, S. Koo, S. Bhuniya, J. S. Kim,
Chem. Soc. Rev. 2015, DOI: 10.1039/C5CS00224A.
[20] D. Andrae, U. Häusermann, M. Dolg, H. Stoll, H. Preuss, Theor. Chim. Acta
1990, 77, 123–141.
[
[
4] S. S. Kelkar, T. M. Reineke, Bioconjugate Chem. 2011, 22, 1879–1903.
5] M. Ali, L. Dondaine, A. Adolle, C. Sampaio, F. Chotard, P. Richard, F. Denat,
A. Bettaieb, P. Le Gendre, V. Laurens, C. Goze, C. Paul, E. Bodio, J. Med.
Chem. 2015, 58, 4521–4528.
[21] a) M. Cossi, V. Barone, B. Menucci, J. Tomasi, Chem. Phys. Lett. 1998, 286,
253–260; b) B. Menucci, J. Tomasi, J. Chem. Phys. 1997, 106, 5151–5158.
[22] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R.
Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nak-
atsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Mont-
gomery Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers,
K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari,
A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M.
Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli,
J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth,
P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B.
Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian 09, revision A.1,
Gaussian, Inc., Wallingford, CT, 2009.
[
6] J. J. Hanthorn, E. Haidasz, P. Gebhardt, D. A. Pratt, Chem. Commun. 2012,
4
8, 10141–10143.
[
7] S. Tasan, O. Zava, B. Bertrand, C. Bernhard, C. Goze, M. Picquet, P. Le Gen-
dre, P. Harvey, F. Denat, A. Casini, E. Bodio, Dalton Trans. 2013, 42, 6102–
6
109.
1
[
8] The value of the coupling constant JP-Pt depends on the nature of the
other ligands coordinated to the Pt center and notably on the nature of
the ligand trans to the P atom. 1
JP-Pt > 3000 Hz values are generally
observed for trans-[P–PtXY–Cl] complexes, whereas lower coupling con-
1
stants ( JP-Pt < 3000 Hz) are reported for trans-[P–PtXY–P] and trans-[P–
PtXY–olefin] complexes, see, for example: a) L. Rigamonti, A. Forni, M.
Manassero, C. Manassero, A. Pasini, Inorg. Chem. 2010, 49, 123–135; b)
A. Massard, V. Rampazzi, A. Perrier, E. Bodio, M. Picquet, P. Richard, J.-C.
Hierso, P. Le Gendre, Organometallics 2012, 31, 947–958; c) J. A. Feducia,
M. R. Gagne, J. Am. Chem. Soc. 2008, 130, 592–599.
Received: November 9, 2015
Published Online: January 5, 2016
Eur. J. Inorg. Chem. 2016, 545–553
www.eurjic.org
553
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim